Help me answer economic questions!!

Anonymous
timer Asked: Sep 29th, 2018
account_balance_wallet $10

Question description

Please help answer questions in the file.

I need them to prepare my midterm πŸ™

THANKS!!!!!

Econ 314 Fall 2018 Instructor: Nusrat Abedin Jimi Midterm Practice Questions (Research Methods) 1) What is internal validity in research? When does a research or an evaluation has strong internal validity? 2) What is external validity in research? When does a research or an evaluation has better external validity? 3) What is identification strategy? Why is it important to have a valid identification strategy? What are some common identification strategies used in applied microeconomics research? 4) Why randomized control trial (RCT) generates a strong internal validity? 5) Suppose a microcredit loan program was randomly assigned among eligible poor households. Due to randomization, the treatment and control group are on average same in terms of both observable and unobservable characteristics before the loan was disbursed. We are interested in the impact of microcredit on earnings. After the program has run for 2 years, we collect information on earnings for both the treatment and comparison units. a) How can you measure the impact of the credit program? 6) Which assumption is the most critical to ensure internal validity of DID models and is the hardest to fulfill? Explain the assumption. What can be done to test the assumption? 7) Which of the following statement is wrong? a) DID model assumes that in absence of treatment, the unobserved differences between treatment and control groups are the same overtime. b) DID requires cohort or panel data (individual level data over time) or repeated crosssectional data (individual or group level). c) As per DID model, comparison groups cannot start at different levels of the outcome compared to the treatment group. d) DID approach removes biases in post-intervention period comparisons between the treatment and control group that could be the result from permanent differences between those groups, as well as biases from comparisons over time in the treatment group that could be the result of trends due to other causes of the outcome. 8) Suppose, your local government is interested to know the impact of a massive road construction program on access to labor markets, with one of the outcome indicators being employment. The road construction program was neither randomly assigned nor was assigned based on an index with a clearly defined cutoff. However, the government is interested to know the causal impact of the program on employment outcomes and hire some economists to know the answer. a) One of the economists suggest that we can simply compare the before-and-after change in employment rates for areas affected by the program. Do you think this will give us the program’s causal impact? Why or why not? b) Another economist suggest that we can compare the employment rates of the areas that received and did not receive the roads program to get the causal impact of the roads on employment. Do you agree with the argument? Why or why not? Explain. c) The third economist suggests comparing the change in employment before and after the program is implemented for individuals living in areas affected by the road construction program to changes in employment in nearby areas (i.e., people living in the neighborhoods near the borders of the affected areas). Which method do you think the third economist is talking about? Will this method help us to know the causal impact of the program on employment? Yes or no? If Yes, explain how. If No, explain why? 9) Difference-in-difference estimation compares the change in the outcome variable (i.e. the pre vs. post estimate) in treatment and control areas. Therefore, a simple difference in difference estimator is given by a) Now consider the following table: Dependent Variable: Employment Rate (%) Before road construction After road construction Areas affected by the 68 75 program Areas not affected by the 70 75 program What is the estimated impact of the program under the DID method? b) To implement the difference-in-difference in a regression framework, we estimate: Where π‘Œπ‘–,𝑑 is the employment rate in an area i at time t; 𝐷𝑖 is the dummy taking a value 1 if an area affected by the road construction program, and 0 otherwise. π‘ƒπ‘œπ‘ π‘‘π‘‘ is an indicator equal to 1 if the time is after intervention and 0 if it is before intervention. 𝛿 is the parameter of interest (it captures the treatment effect or the effect of the road construction program). Using data, we estimate the regression and get the following result. π‘ŒΜ‚π‘–,𝑑 = 0.04 + 0.36 𝐷𝑖 + 0. 13 π‘ƒπ‘œπ‘ π‘‘t + 0.21 𝐷𝑖 βˆ— π‘ƒπ‘œπ‘ π‘‘π‘‘ (0.03) (0.23) (0.06) (0.12) where the standard errors are reported in parenthesis. Based on the estimates, will you report to the government that the road construction program has a significant impact on employment rate? 10) How does the instrumental variable method help us to obtain an unbiased estimator in OLS and deal with endogenous X variable? What are the requirements for a valid instrument? 11) Suppose, we are interested in the impact of schooling on earnings. Our regression equation is π‘’π‘Žπ‘Ÿπ‘›π‘–π‘›π‘”π‘  = 𝛽0 + 𝛽1 π‘ π‘β„Žπ‘œπ‘œπ‘™π‘–π‘›π‘” + Ο΅ Our parameter of interest is Ξ²1. However, running a simple OLS regression using observational data like this might lead to biased estimation of returns to schooling because of different confounding factors (schooling is endogenous. i.e., there might be omitted variables that are correlated to schooling which lead to omitted variable bias). Now, we plan to use instrumental variable method to estimate the returns to schooling (impact of schooling on earnings). For that purpose, we need to have another variable that can be used as an instrument of the endogenous variable schooling. a) Someone suggests that we can use mother’s education as an instrument for schooling. Do you think mother’s education is a good and valid instrument for schooling? Explain your reasoning. b) Another person suggests using β€˜distance from home’ as an instrument for schooling. Do you think distance from home serves as a valid instrument for schooling? Explain your reasoning. 12) Poor households often deal with their day to day financial decisions with very limited numbers of financial instruments (i.e, less access to loans, savings, insurance etc.). This limited access eventually leads to lower earnings and living standards. Suppose, there is an NGO that ran a microcredit program and offered small loans to poor households of 20 villages of Srilanka. Access to loans/microcredit was assigned randomly. However, not all households who had the loan access actually took it, which implies that there were not 100 percent compliance. The program officials are interested to know the impact of small loans on earnings of poor households. The relation between credit and earnings is as follows: π‘™π‘œπ‘” (π‘’π‘Žπ‘Ÿπ‘›π‘–π‘›π‘”π‘ ) = 𝛽0 + 𝛽1 π‘π‘Ÿπ‘’π‘‘π‘–π‘‘ + πœ– Where π‘’π‘Žπ‘Ÿπ‘›π‘–π‘›π‘”π‘  is monthly earnings of the household and credit is a dummy variable which takes a value of 1 if the household took loan and 0 otherwise. a) What is our parameter of interest? b) An economist argues that running the above mentioned simple OLS regression to estimate the impact of credit will lead to a biased estimation. Why do you think that his argument is reasonable? c) The economist suggests that we should estimate the impact of credit on earnings using an instrumental variable method where the loan offer is used as an instrument for credit. Why the loan offer is a valid and strong instrument of credit? Explain. 13) Explain the statement β€œRegression Discontinuity Design has strong internal validity but limited external validity”. 14) How does the propensity score matching (PSM) method construct an artificial comparison group? What is the most critical assumption of the propensity score matching (PSM) method.

Tutor Answer

Dmitry
School: UCLA

checki...

flag Report DMCA
Review

Anonymous
Wow this is really good.... didn't expect it. Sweet!!!!

Similar Questions
Hot Questions
Related Tags

Brown University





1271 Tutors

California Institute of Technology




2131 Tutors

Carnegie Mellon University




982 Tutors

Columbia University





1256 Tutors

Dartmouth University





2113 Tutors

Emory University





2279 Tutors

Harvard University





599 Tutors

Massachusetts Institute of Technology



2319 Tutors

New York University





1645 Tutors

Notre Dam University





1911 Tutors

Oklahoma University





2122 Tutors

Pennsylvania State University





932 Tutors

Princeton University





1211 Tutors

Stanford University





983 Tutors

University of California





1282 Tutors

Oxford University





123 Tutors

Yale University





2325 Tutors